My recent focus is the the Solar neighborhood with the goal of
using large stellar kinematic and abundance surveys to unravel structure in the Galaxy.
Nearby stars show variations in the local velocity distributions.
We figured out a way to relate open clusters abundances to migration models.
We ran simulations of large numbers
of test stars in parallel on a graphics card inside an N-body simulation.
With Jamie and Micaela we
found a connection between gaps in local disk velocity distributions,
kinks or discontinuities in spiral arms, and Lindblad resonances.
Interference between spiral density waves gives interesting phenomena including
waves of star formation, and slow moving
interference peaks that can mediate stellar migration.
We proposed a resonant heating model for the X- or peanut-shape
in the Milky Way's galactic bulge.
The outer galaxy is likely warped, spiral and has large velocity gradients due
to external perturbations (such as from a dwarf galaxy) and has
been mixed in abundance distribution.
I previously used dynamical models to place constraints
on the dark matter distribution in galaxies.
Review talk at Yale 2015 pptx
Click here for movie of tracer particles being perturbed by spiral structure (on the shearing sheet)!
We find that phase wrapping of epicyclic
perturbations could explain velocity gradients
recently interpreted (misinterpreted?) as bending waves and breathing modes in the Galaxy.
It is possible that the recent pericenter of Sagittarius dwarf galaxy 1 Gyr
ago is responsible for the epicyclic perturbations.
Alex's paper!
arxiv
,
Phase Wrapping of Epicyclic Perturbations in
the Wobbly Galaxy
A resonant heating model for X- or Peanut shaped galactic bulges
,
ADS
Inspired by the recent discovery of a X- or peanut shape to
our Galaxy bulge, I re-examined the strength of
the 2:1 vertical resonance (2 vertical oscillations per rotation
in the bar frame).
I was surprised to find that the vertical resonance in simulations
is quite narrow and is drifting outwards in the simulations. This
implied that my
old resonance capture model
needed correction.
The new model is a resonance heating model :
Stars in the midplane, as they enter the resonance, are lifted
to high inclinations to orbits just inside the resonance separatrix, distant
from the banana shaped periodic orbits but supporting the peanut shape rotating
with the bar.
The X-shape arises from the shape of orbits of the few stars
that are currently in or near resonance.
Stars exterior to resonance would not support the peanut shape.
The X-shape stars would have metallicities
like the disk exterior to the resonance.
The Hamiltonian model can be used to place constraints on the mass distribution
in the central 1.5 kpc of the Milky Way Galaxy.
Talk given at CITA Oct 2013
pptx
Below are two movies of N-body simulations,
one showing a bar that buckles (on left), the other
without buckling (on right). These movies show face-on and edge-on views of
barred galaxies.
Both movies show that the peanut shape becomes more extended
as the bar slows down.
Quicktime 7 movies:
Mp4 movies:
These are showing Galmer simulations and the movies were made by Ivan Minchev.
Review Talk May/June, 2013 at the IAU Symposium in Lijiang,
on spiral and bar galactic structure and how they affect velocity distributions
pptx
Dissection of an N-body galactic disk exploring structure in phase space
associated with bar and spiral waves (with Micaela and Jamie)
Structure in phase space associated with spiral and bar density waves in an N-body hybrid galactic disc (MNRAS),
astroph
For decades there have been two dominant scenarios for galactic spiral
structure: a modal view and a transient view. Oddly we don't interpret
our N-body
simulations in terms of either view. Instead we see multiple waves, with
interference between them giving transient phenomena. I think
the spiral waves are coupled to the bar in this simulation and so are not
self-excited transients or modes of the disk. Check out the movies:
The first of these movies is pretty ordinary (except for Jamie's garish color scheme). The second of these movies presents a polar view of spiral structure.
At the time we had never seen an animated polar view of an N-body spiral disk, but lately other groups have been presenting them for similar reasons;
it is easier to see if spiral patterns are slow in the outer parts
of the galaxy and fast in the inner region.
Here the density distribution in polar coordinates shows how two and three arm waves
intefere in a galaxy disk. Discontinuities in spiral structure such
as armlets, kinks and bends are related to different ranges spanned by
different waves.
The most interesing coupling is that
between the bar, a lopsided motion and a three armed wave.
In the Milky Way the three-armed wave should be near the solar neighborhood
(Why? because in our simulation I see the 3 armed structure going between
Inner Lindblad Resonance (ILR) and Outer Lindblad Resonance and it has
ILR at the bar's corotation resonance).
Oddly I find that there have been no attempts to fit Galactic spiral structure
observations to combinations of two and three armed models. Almost everybody
assumes a bi-symmetric (2+4) model even though the outer Galaxy is distinctly and
uncontroversially lopsided (HI observations)-- and so there should be a three armed structure coupled to the bar.
This work makes a few predictions 1) there is a three armed structure
near the solar neighbhorhood, 2) gaps in local velocity distributions will
be found all over the Galaxy, 3) there are transient bursts of star formation
caused by interference between patterns and appearance of short lived peaks
(this is an altnerative to the steady progression of a single spiral density wave).
Radial mixing in the outer Milky Way disk caused by a low mass orbiting satellite
astroph ,
ADS
Micaela's ring galaxy simulations. We integrate
100,000 particles simultaneously on a graphics card. Collisionless
particles if they start out with low velocity dispersion show morphology
remarkably similar to outer rings in galaxies (ring structure
previously interpreted as long lived might instead be transient). We explore
constraints on secular evolution of bars and length of time since
bar formation.
PDF (this copy with higher resolution figures than that on astroph) ,
The morphology of galactic rings exterior to evolving bars
Micaela's simulation of ring formation
by a galactic bar
The simulation of 100,000 massless disk stars integrated
perturbed by a galactic bar. This ran in 60 seconds on the GPU. The
outer ring has morphology similar to R1'R2 pseudo-ringed galaxies.
Our Spitzer imaging survey of brightest cluster galaxies (yes, they can host star formation),
paper II ,
paper I
Older work includes a nice algorithm to estimate the gravitational potential from an image and an estimate of the gas inflow rate along the bar in the galaxy NGC7479.